Software Engineer

Experience level: 3–5 years

Location: [Hybrid –Bangalore]

Primary Skills - Java Full Stack and AI Automation

Role Summary

We are looking for an AI Automation Engineer to design, build, and operate AI-powered automations that remove manual effort from business and engineering workflows. This is a hands-on build role: you will take ambiguous, human-in-the-loop processes and turn them into reliable, monitored, production-grade automations using LLMs, agentic frameworks, APIs, and workflow orchestration tools.

You will work closely with process owners, data engineers, and platform teams to identify high-value automation candidates, prototype quickly, and harden what works into supportable services.

Key Responsibilities

• Build and deploy end-to-end AI automations and agentic workflows that integrate LLMs with internal systems, APIs, and data sources.
• Design and implement RAG (retrieval-augmented generation) pipelines: chunking, embedding, vector search, re-ranking, and grounded response generation.
• Develop tool- and function-calling integrations so agents can act on systems of record (ticketing, ERP, document stores, internal APIs), including building MCP servers to expose internal systems as agent tools.
• Partner with business stakeholders to map current-state processes, quantify manual effort, and prioritize an automation backlog by ROI and feasibility.
• Write production Python services and jobs — API clients, data transformations, schedulers, event-driven handlers — with proper logging, retries, and error handling.
• Build prompt and model evaluation harnesses: golden datasets, regression tests, accuracy and hallucination checks, and pre-deployment quality gates.
• Instrument automations with observability and tracing so failures, cost spikes, latency regressions, and quality drift are detected early.
• Implement human-in-the-loop checkpoints, approval steps, and safe fallbacks for workflows where full autonomy is inappropriate.
• Optimize for cost and latency through model selection, caching, batching, prompt compression, and routing between larger and smaller models.
• Document architecture, runbooks, and handover material; support and iterate on automations after launch.
• Follow security, privacy, and responsible-AI requirements around data handling, access control, PII, and model usage.

Required Qualifications

• 3–5 years of professional software engineering, data engineering, or automation engineering experience, including at least 1 year building with LLMs or AI services.
• Strong Python skills — async programming, API integration, testing, packaging, and clean, maintainable code.
• Practical experience with LLM APIs (e.g. Anthropic Claude, OpenAI, Gemini, or Bedrock) in a real application, not just experimentation.
• Hands-on experience with at least one agent or orchestration framework (LangChain, LangGraph, LlamaIndex, CrewAI, Semantic Kernel, AutoGen, or equivalent in-house tooling).
• Demonstrated experience building RAG systems and working with a vector store (pgvector, Pinecone, Weaviate, Qdrant, Milvus, OpenSearch, or similar).
• Solid grasp of prompt engineering and context design, plus structured output techniques (JSON schemas, tool/function calling, validation and repair).
• Familiarity with the Model Context Protocol (MCP) — building MCP servers to expose internal systems as agent tools, and integrating them into agent applications.
• Strong REST/GraphQL API integration skills, including OAuth and token-based authentication, webhooks, pagination, and rate-limit handling.
• Working knowledge of SQL and data modeling; comfortable joining and cleaning data across systems.
• Experience with Git-based workflows, code review, and CI/CD.
• Cloud experience on AWS, Azure, or GCP — compute, serverless functions, queues, secrets management, and managed AI services.
• Containerization fundamentals with Docker.
• Proven ability to work directly with stakeholders: eliciting requirements, demoing prototypes, and managing expectations.

Preferred Qualifications

• Experience with workflow automation platforms: n8n, Zapier, Power Automate, Airflow, Temporal, Prefect, or Dagster.
• Background in RPA (UiPath, Automation Anywhere, Blue Prism) and experience modernizing RPA-era automations with AI.
• Experience with LLM observability and evaluation tooling (LangSmith, LangFuse, Arize, Braintrust, Ragas, or DeepEval).
• Multi-agent system design — planner/executor patterns, supervisor architectures, delegation, and shared state.
• Document intelligence and multimodal work: OCR, PDF parsing, table extraction, vision model usage.
• Frontend familiarity (React.js, Next.js, TypeScript) for building internal tools and demos.
• Infrastructure-as-code (Terraform, CloudFormation) and Kubernetes exposure.
• Experience with enterprise governance concerns: AI risk review, audit trails, data residency, and access-scoped retrieval.
• Bachelor's degree in Computer Science, Engineering, or a related field, or equivalent practical experience.